Maybe looping back a little to Aaron's
original question which I'm
guessing is from Shannon: we don't have a way to measure " Number of
editors who contribute 1 edit per month" as we don't have ways to
accurately identify people who use multiple accounts, IPs, etc. We do have
ways to track number of unique accounts, but that's different from number
of unique editors. Although it would be nice to know how many people edit
the projects on any given month, that's impossible to know, although maybe
Nuria and the other analytics folks would have some ideas on how to get an
approximation.
Pine
On Tue, Apr 11, 2017 at 9:48 PM, Pine W <wiki.pine(a)gmail.com> wrote:
If we're going to have a conversation about
terminology, I would like
to drop the terms "active editor" and "highly active editor" and
replace
them with "5+ edits per month" and "100+ edits per month". There are
multiple ways of measuring productivity, and I'm wary of the amount of
prominence that's given to the number of edits as the primary metric of
productivity. Also, I don't think it's clear to analytics nebwbies that
"active editor" is a term with a specific definition rather than a general
description of people who edit "actively" (whatever that means). I'm fine
with using 5+ edits per month and 100+ edits per month as measures of
productivity, but I would prefer to drop the terms "active editor" and
"very active editor". I'd also like to see more prominence given to other
metrics such as bytes changed and logged non-edit actions.
Pine
On Tue, Apr 11, 2017 at 10:22 AM, Erik Zachte <ezachte(a)wikimedia.org>
wrote:
> Aaron,
>
>
>
> Yeah my analogy is arguably imprecise.
>
> And for your analogy, you assume that the public astronomy database is
> guarded Nupedia style, with credentials. Could be, explicit mention of this
> assumption would resolve ambiguity ;-)
>
> > Our licensing asserts that they must be attributed.
>
>
>
> Sure these people who did one edit must be attributed whenever the
> page they edited is published somewhere else.
>
> But do we ever do that for real these days? Seems like a dead clause
> from a distant past, expect for our onwiki history page.
>
>
>
> Also giving credit is something else than counting, and publishing
> that count as some meaningful metric (not saying that you want to do that,
> but others will find the factoid and run with it)
>
> We can discuss semantics. But when a person writes one word a year we
> wouldn't call that person a 'writer', do we?
>
> Words lose their meaning if their definition is stretched in extremo,
> beyond common sense, beyond what any audience assumes those words mean.
>
>
>
> Long ago we found that a huge amount of registered users made not even
> one edit.
>
> One explanation might be that many people habitually sign up, just out
> of habit. Or that they want to tweak the UI (e.g. red links in
> preferences).
>
>
>
> My point: count as you like, but could we avoid using a term with so
> many connotations for these edge cases, so as not to confuse people even
> more about our metrics?
>
>
>
> Erik
>
>
>
>
>
> *From:* Analytics [mailto:analytics-bounces@lists.wikimedia.org] *On
> Behalf Of *Aaron Halfaker
> *Sent:* Tuesday, April 11, 2017 16:55
>
> *To:* A mailing list for the Analytics Team at WMF and everybody who
> has an interest in Wikipedia and analytics.
> *Subject:* Re: [Analytics] Fwd: follow-up on editors
>
>
>
> Erik,
>
>
>
> I appreciate pushing back on just looking for bigger metrics, but
> there's something more important when it comes to measuring people who
> contribute at least a little bit. Our licensing asserts that they must be
> attributed. After all, they have contributed something.
>
>
>
> Also, for your astronomy comparison, this would be more like saying
> that anyone who contributes to publicly recorded astronomy observations is
> an astronomer -- even if they have only done so once. In my estimation,
> that doesn't sound crazy. Your comparison to "looking at the night
sky" is
> a lot more like reading Wikipedia.
>
>
>
> -Aaron
>
>
>
> On Tue, Apr 11, 2017 at 6:35 AM, Erik Zachte <ezachte(a)wikimedia.org>
> wrote:
>
> About 'Number of editors who contribute 1 edit per month?'
>
>
>
> I'm hoping we're not going that use that number for our next
> fundraiser ;-)
>
> The more inclusive our numbers are, the less meaningful, bordering on
> alternative facts.
>
>
>
> A person with one edit in any given month is as much an editor as a
> person who looks at the night sky a few times a year is an astronomer.
>
> We have billions of those on this planet!
>
>
>
> Erik
>
>
>
>
>
> *From:* Analytics [mailto:analytics-bounces@lists.wikimedia.org] *On
> Behalf Of *Neil Patel Quinn
> *Sent:* Friday, March 31, 2017 23:06
> *To:* A mailing list for the Analytics Team at WMF and everybody who
> has an interest in Wikipedia and analytics.
> *Subject:* Re: [Analytics] Fwd: follow-up on editors
>
>
>
> Funny story: I noticed that Aaron's graph has the 1-month new editor
> retention on enwiki at about 7%, while I had recently done some
> queries
>
<https://github.com/wikimedia-research/2017-New-Editor-Experiences/blob/master/analysis.ipynb>
> that put it a little under 4%.
>
> It turns out I made an error in my Unix timestamp math, and I was
> looking at the *12 hour *new editor retention rate. It'll be
> interesting to see if the ranking of wikis by retention changes
> significantly when I correct that.
>
>
>
> On Wed, Mar 29, 2017 at 2:15 PM, Aaron Halfaker <
> ahalfaker(a)wikimedia.org> wrote:
>
>
> *https://commons.wikimedia.org/
>
<https://commons.wikimedia.org/>wiki/File:Enwiki.monthly_user_retention.survival_proportion.svg*
>
>
>
> On Wed, Mar 29, 2017 at 4:14 PM, Aaron Halfaker <
> ahalfaker(a)wikimedia.org> wrote:
>
> Here's a graph of the retention rates of new editors in English
> Wikipedia.
>
>
>
>
>
>
> _______________________________________________
> Analytics mailing list
> Analytics(a)lists.wikimedia.org
>
https://lists.wikimedia.org/mailman/listinfo/analytics
>
>
>
>
> --
>
> Neil Patel Quinn
> <https://meta.wikimedia.org/wiki/User:Neil_P._Quinn-WMF>, product
> analyst
> Wikimedia Foundation
>
>
> _______________________________________________
> Analytics mailing list
> Analytics(a)lists.wikimedia.org
>
https://lists.wikimedia.org/mailman/listinfo/analytics
>
>
>
> _______________________________________________
> Analytics mailing list
> Analytics(a)lists.wikimedia.org
>
https://lists.wikimedia.org/mailman/listinfo/analytics
>
>
_______________________________________________
Analytics mailing list
Analytics(a)lists.wikimedia.org